Why Doesn’t Capital Flow to Where Returns Are High?

The growth of global imbalances—implying ‘uphill’ flows of capital from emerging economies with relatively low capital intensity and high rates of return to developed economies with mature capital stocks and relatively low rates of return—has been one of the most important economic developments of the past decade. While the size of these imbalances have shrunk since the …

Se vuoi ricevere le principali notizie riguardanti gli ETF e gli ETC iscriviti alla Nostra Newsletter settimanale gratuita. Clicca qui per iscriverti gratuitamente.

on set of the financial crisis, capital continues to flow ‘uphill’ from poor economies to rich ones. In understanding the causes of these ‘uphill’ flows, three key questions arise: (i) Are poor countries’ returns actually higher than rich ones?; (ii) Does capital flow from countries where returns are low to countries where returns are high?; and, (iii) If not, why not? We find that: (i) Rates of return on physical capital are significantly higher in poor economies than in rich economies (to a degree that exceeds any reasonable premium for the additional riskiness of those returns); (ii) Capital does not typically flow from lowreturn economies to high-return economies; and, (iii) The existence of ‘uphill’ capital flows from poor countries with high returns to rich countries with low returns appears primarily to be due to differences in demographics between emerging and developed economies, and the effect of higher trend growth on aggregate savings behaviour. The key insight for investors is that, in a world without perfect capital mobility and with capital flows being driven by factors that— although rational from the perspective of the saver—are unrelated to return differentials, the opportunities for ‘unconstrained’ investors (i.e., those who are in a position to take the other side of those flows) to invest where returns are highest are likely to be greater.

Why Doesn’t Capital Flow to Where Returns Are Highest?

The Lucas Paradox: Why Doesn’t Capital Flow from Rich to Poor Countries?

This week we summarise the findings of Global Economics Paper No. 205 (November 12, 2010). Economic theory suggests that, in the absence of full capital market integration, rates of return should be higher in poor countries than in rich ones and that, reflecting these differences, capital should flow from rich to poor countries. In 1990 Robert Lucas, the Nobel prize-winning economist, posed the question “Why doesn’t capital flow from rich to poor countries?”, and various explanations have since been proposed as to why this does not appear to be the case. The paradox can be usefully split into two questions: “Are rates of return higher in poor countries?” and “Do capital flows systematically respond to differences in rates of return?”, and proposed explanations typically address one of these two questions. The first group of explanations refers to differences in fundamentals affecting the production structure of economies, which have the effect of reducing or even reversing rate of return differentials across rich and poor countries. One possibility is that differences in human capital fully counteract other factors driving differences in profitability. The second group of explanations relies on international capital market imperfections. For instance, it may be that badly-defined property rights in cross-border investments entirely offset the incentive to invest where returns are highest.1

Returns Are Higher in Poor Economies, But Aggregate Capital Flows Are Not ‘Return Chasing’

Using a new database of ex-post rates of return on capital for the 12 largest economies in the world, covering threequarters of global output and spanning more than onequarter of a century, we find considerable and persistent differences in ex-post rates of return on capital. The differences we find are unlikely to be due to measurement error: the data are based on national accounts data for the private non-financial sector (where profitability and capital stocks are most accurately measured), and have been adjusted for any differences in methodology that exist across countries. Nor can the differences be explained by the distinction between ex-post and ex-ante returns—we find evidence of predictability in the crosscountry variation of ex-post returns—or by differences in the degree of risk attached to returns. Using panel-data analysis, we find that cross-country differences in rates of return are positively correlated with growth in GDP per capita and negatively correlated with the level of GDP per capita, as standard (neo-classical) theory would predict (Figure 1). However, we find no systematic response of overall cross-country capital flows to these differences (Figure 2). This failure does not appear to be due to the distorting effects of reserve accumulation. Within overall capital flows, we separately consider the response of net inflows of foreign direct investment, equity investment and bond investment—the sub-components that one would expect to be most strongly ‘return chasing’—but find no systematic response to differences in the return on capital here either. Our findings imply a ‘yes’ in answer to the question of whether rates of return are higher in low-income countries but a strong ‘no’ to the question of whether capital flows systematically respond to differences in the rates of return.

Large and Persistent Differences in Rates of Return
The core of our analysis is based on a new cross-country database of returns on physical capital (see Global Paper No. 205 for more details). Using this database, we find:

– Large cross-country differences in rates of return: Figure 3 displays the mean return on capital measure, rK , for each country from 1981 and 2008. Figure 4 displays the rates of return on capital for Europe (Germany, UK, France, Italy and Spain), Asia (Japan, China, Korea and India) and the US. The variation in mean cross-country returns has been substantial, from a high of 22% for India to a low of 8% for Spain. Even geographically-neighbouring countries saw substantially different rates of return: France’s return on capital averaged 4% less than Italy’s over the 28-year sample.2

The ranking of cross-country rates of return on capital has changed significantly over the 28-year sample. In Asia, for instance, Japanese and Korean returns were very high in the 1980s but subsequently fell, while China’s ROC increased significantly in the 2000s.

Exchange rate changes matter but have not smoothed out the cross-country returns: Exchange rate fluctuations matter for investors, with interest rate parity implying convergence in (ex-ante) rates of return only once expected exchange rate fluctuations are also taken into account. However, the difference in mean returns across countries (as measured by the cross-sectional standard deviation) is even higher in US-Dollar terms than in local-currency terms. Moreover, long-run exchange rate moves appear to be unrelated to rate of return differentials.

The cross-country differences in rates of return cannot be explained by differences in risk: Figure 5 plots the mean results for rK against their respective standard deviations. A positive relation exists between the mean and the volatility of returns, as theory would imply, but the relation is not strong (the R-squared is 0.28). Moreover, if exchange rate effects are also taken into account, the positive relation between mean and standard deviation disappears entirely. Nor do these differences appear to be due to cross-country variation in property rights: one of the advantages of using a return on capital measure based on reported profits (rather than ‘top-down’ estimates of the profits share) is that it is more likely to reflect the true return available to investors (because appropriated profits, by
their nature, are unlikely to be reported in the national accounts).

Rates of Return are Higher in Low-income, Fast-Growing Economies

Table 1 displays the results of a series of panel-data estimations of the relation between the return on capital, rK , and a selection of explanatory variables. We find that:

Rates of return on capital are higher in rapidlygrowing economies—strong growth boosts profitability. Regression 1.1 suggests that there is a strong positive relation between the return on capital (rK) and annual real GDP growth, with every onepercent rise in long-term growth boosting the ROC by 1.4ppt. This effect operates via GDP per capita rather than population growth: Regression 1.2 implies a strong positive relation between ROCs and GDP per capita growth rates but none with population growth.

The return on capital is typically higher in poor countries.
Regression 2.3 implies a strong negative relation between the return on capital and the level of GDP per capita. Every 10ppt convergence with US GDP per capita has, all else equal, historically reduced the return on capital by close to one percent.

Capital Flows do Not Respond to Differences in Rates of Return

Economic theory implies that capital flows should respond to differences in rates of return, with countries with relatively low rates of return exporting capital to countries with relatively high rates of return. In Figure 2, we plotted the mean difference between country i’s rate of return and the global rate of return against the average capital inflow (expressed as a percentage of GDP). Based on average data, there is no relation (either positive or negative), in contrast to the standard theoretical prediction. The regressions set out in Table 2 explore the relation between capital inflows and relative rates of return more formally:

– Capital does not flow to countries where returns are highest. Regression 2.1 implies no relation between aggregate capital inflows (the negative of the current account balance) on the difference between the return on capital in country i and the global return on capital. Indeed, the coefficient has been negative (but not significant). In 2.2 we carry out the same regression using post-tax rather than pre-tax rates of returns, with similar results. The absence of a significant relation is robust to different lag structures in the return differential and across time (splitting the 1981-2008 sample into two 14-year periods, we find no significant relation in either period).

– Private-sector capital flows (i.e., excluding reserve accumulation) also appear to be largely unresponsive to return differentials. In the past year and a half, private-sector capital flows into EM economies have increased significantly, offset by public-sector reserve accumulation (a development we have highlighted in our FX research—see “The ‘Wall of Money’ to EM”, Global FX Monthly Analyst, October 2010). In general, however, we find that private-sector capital flows have not been ‘return chasing’. Regressions 2.3-2.5 focus on the subcomponents of capital flows—foreign direct investment, net equity investment and net bond investment—where theory provides the strongest a priori expectation of a systematic response to return differentials. In each case the coefficient is positive— as theory would predict—but not significant.

– The failure of capital to flow to where returns are highest is not due to mis-measurement of ROC levels. Although we have taken care to ensure the cross-country comparability of our ROC measures, some scope for mis-measurement will always remain, particularly when comparing ROC levels. To account for the possibility of mis-measurement, Regression 2.6 repeats 2.2 but allows for country-specific fixed country effects. The results are very similar, implying that capital flows do not respond to relative increases in the return on capital over time.

Intergenerational Savings Dynamics and Demographics Dominate ‘Return Chasing’ Flows

One can identify a number of alternative factors—i.e., other than differences in rates of return—that could be driving cross-country capital flows. An economy’s current account is equal to its ‘net’ saving (saving less investment), so a tendency to save more across an economy will translate into pressure for current account surpluses and a flow of capital to other countries. Standard theory implies that it is the rate of return on capital that determines the equilibrium between those providing savings and those making investments. But other factors may induce some countries to save more than others:

– Cross-country demographic differences: Because people’s savings behaviour varies throughout the different stages of their lives, the relative age structure of an economy can play an important role in determining their borrowing and lending to the rest of the work (see Global Economics Paper No. 202, August 12, 2010).

– Higher trend growth can boost net saving: One would typically expect rapidly growing economies to attract capital inflows but high trend growth can also boost net saving. In rapidly growing economies, the current generation of workers will typically earn much higher levels of income than the current generation of retirees will have done. Rapid growth can increase the aggregate level of saving in an economy if the savings of the (relatively rich) young cohorts is greater than the dissaving of (relatively poor) retirees.

– Differences in financial development: It is argued by some economists that low levels of financial development in emerging economies lead emerging market investors to seek more trustworthy savings vehicles in the mature financial markets of the developed world, leading to ‘uphill’ capital flows.

– Cross-country differences in the degree of risk aversion: Some countries may have a higher level of precautionary saving than others if future income is less certain. For instance, one suggestion is that the high household saving rates of emerging economies reflect a high level of individual risk, related to health costs, retirement and the financing of education, as a result of low levels of social protection.

We consider the effects of demographics, economic growth and financial development on net capital flows (Table 3 presents panel-data analysis of the relation between capital inflows and each of these variables). Our findings suggest that:

Countries exhibit capital outflows when the working age population is relatively high. Regression 3.1 suggests that there are net capital outflows when a relatively high proportion of the population is of working age but the coefficient just fails to be significant at the 10% threshold.

There is a strong relationship between net outflows and high trend growth. The size of the coefficient in Regression 3.2 suggests that every one-percent difference in trend growth is matched one-for-one by a 1%-of-GDP increase in capital outflows in the long run.

There is no discernible link between capital flows and our measure of financial development (Regression 3.3).
Regressions 3.4 explores how each of these variables interact when included together, based on annual data. We find strong evidence that higher trend per capita growth results in capital outflows and weak evidence that a relatively high working age ratio results in capital outflows.

To summarise the results, we find that aggregate net capital flows do not respond to cross-country differences in rates of return. Rather than being ‘return chasing’, long-term net capital flows appear to be largely dictated by internal savings and investment dynamics (which, in turn, are driven by intergenerational savings dynamics and demographics).

Investment Implications: Better Opportunities for ‘Unconstrained’ Investors

There is a wide body of opinion that ‘uphill’ capital flows from emerging economies with high internal rates of return to developed economies with low rates of return played an important part in the formation of the 2007/2008 financial crisis (see, for instance, “The Savings Glut, the Return on Capital and the Rise in Risk Aversion”, Global Economics Paper No. 185, May 2009). If one subscribes to the view that the ‘savings glut’ had an important role in the formation of the financial crisis, then understanding why capital does not flow from countries with low rates of return to countries
with high rates of return is key to understanding the formation of the financial crisis itself. Indeed, the ‘savings glut’ that preceded the 2007/08 financial crisis can be viewed as a special case of a much wider phenomenon, with the principal difference being that the emerging markets that are currently progressing through a period of rapid transition (in particular, the BRICs economies) are much larger than the transition economies of the past.
This also has important implications for investors: in a world without perfect capital mobility, ‘uphill’ capital flows from fast-growing, high-return economies to slowgrowing, low-return economies contribute to and perpetuate the existence of cross-country return differentials. If capital were flowing in the ‘right’ direction, these flows would tend to smooth out return differentials over time. With capital flows that are driven by factors that—although rational from the perspective of the saver—are unrelated to return differentials, the opportunities for ‘unconstrained’ investors (i.e., those
who are in a position to take the other side of those flows) to direct capital to where returns are highest are likely to be greater.
We emphasise ‘likely’ to be greater because one can still overpay for the underlying assets that produce relatively high returns. But equity returns have also typically been higher in countries with relatively high returns on physical capital—consistent with the view that ‘uphill’ capital flows have led to underinvestment in EM financial capital (as well as physical capital—Figure 6). Moreover, we find little evidence that return differentials are typically reflected in stock market valuations (Figure 7). We believe that the market implications of this analysis are primarily of a medium-to-long-term significance, rather than implying that EM equities are necessarily ‘cheap’ today. However, it is also the case that our equity strategists do not view the majority of EM markets to be currently overvalued.
In past research we have emphasised the opportunities that the BRICs and other EM economies present for both portfolio and direct investment. The findings set out in this paper provide a new perspective on those opportunities. They suggest a mechanism whereby (riskadjusted) return differentials can and do persist.

Kevin Daly

Equity Risk and Credit Premiums
In October, our ECP was 374bp higher than the most recent trough in November 2008. The US ERP has decreased by 51bp since its most recent peak in mid-October, due to the rise in real bond yields.

Global product; distributing entities
The Global Investment Research Division of Goldman Sachs produces and distributes research products for clients of Goldman Sachs,and pursuant to certain contractual arrangements, on a global basis. Analysts based in Goldman Sachs offices around the world produce equity research on industries and companies, and research on macroeconomics, currencies, commodities and portfolio strategy. This research is disseminated in Australia by Goldman Sachs & Partners Australia Pty Ltd (ABN 21 006 797 897) on behalf of Goldman Sachs; in Canada by Goldman Sachs &Co. regarding Canadian equities and by Goldman Sachs & Co. (all other research); in Hong Kong by Goldman Sachs (Asia) L.L.C.; in India by Goldman Sachs (India) Securities Private Ltd.; in Japan by Goldman Sachs Japan Co., Ltd.; in the Republic of Korea by Goldman Sachs (Asia) L.L.C., Seoul Branch; in New Zealand by Goldman Sachs & Partners New Zealand Limited on behalf of Goldman Sachs; in Russia by OOO Goldman Sachs; in Singapore by Goldman Sachs (Singapore) Pte. (Company Number: 198602165W); and in the United States of America by Goldman Sachs &Co. Goldman Sachs International has approved this research in connection with its distribution in the United Kingdom and European Union. European Union: Goldman Sachs International, authorized and regulated by the Financial Services Authority, has approved this research in connection with its distribution in the European Union and United Kingdom; Goldman Sachs & Co. oHG, regulated by the Bundesanstalt fur Finanzdienstleistungsaufsicht, may also distribute research in Germany.
General disclosures
This research is for our clients only. Other than disclosures relating to Goldman Sachs, this research is based on current public information that we consider reliable, but we do not represent it is accurate or complete, and it should not be relied on as such. We seek to update our research as appropriate, but various regulations may prevent us from doing so. Other than certain industry reports published on a periodic basis, the large majority of reports are published at irregular intervals as appropriate in the analyst’s judgment.
Goldman Sachs conducts a global full-service, integrated investment banking, investment management, and brokerage business. We have investment banking and other business relationships with a substantial percentage of the companies covered by our Global Investment Research Division. SIPC: Goldman, Sachs & Co., the United States broker dealer, is a member of SIPC (http:// www.sipc.org).
Our salespeople, traders, and other professionals may provide oral or written market commentary or trading strategies to our clients and our proprietary trading desks that reflect opinions that are contrary to the opinions expressed in this research. Our asset management area, our proprietary trading desks and investing businesses may make investment decisions that are inconsistent with the recommendations or views expressed in this research.
We and our affiliates, officers, directors, and employees, excluding equity and credit analysts, will from time to time have long or short positions in, act as principal in, and buy or sell, the securities or derivatives, if any, referred to in this research.
This research is not an offer to sell or the solicitation of an offer to buy any security in any jurisdiction where such an offer or solicitation would be illegal. It does not constitute a personal recommendation or take into account the particular investment objectives, financial situations, or needs of individual clients. Clients should consider whether any advice or recommendation in this research is suitable for their particular circumstances and, if appropriate, seek professional advice, including tax advice. The price and value of investments referred to in this research and the income from them may fluctuate. Past performance is not a guide to future performance, future returns are not guaranteed, and a loss of original capital may occur. Fluctuations in exchange rates could have adverse effects on the value or price of, or income derived from, certain investments.
Certain transactions, including those involving futures, options, and other derivatives, give rise to substantial risk and are not suitable for all investors. Investors should review current options disclosure documents which are available from Goldman Sachs sales representatives or at http://www.theocc.com/publications/risks/riskchap1.jsp. Transactions cost may be significant in option strategies calling for multiple purchase and sales of options such as spreads. Supporting documentation will be supplied upon request.
All research reports are disseminated and available to all clients simultaneously through electronic publication to our internal client websites. Not all research content is redistributed to our clients or available to third-party aggregators, nor is Goldman Sachs responsible for the redistribution of our research by third party aggregators. For all research available on a particular stock, please contact your sales representative or go to www.360.gs.com.
Disclosure information is also available at http://www.gs.com/research/hedge.html or from Research Compliance, 200 West Street, New York, NY 10282. No part of this material may be (i) copied, photocopied or duplicated in any form by any means or (ii) redistributed without the prior written consent of The Goldman Sachs Group, Inc. © Copyright 2010, The Goldman Sachs Group, Inc. All Rights Reserved.

The World in a Nutshell

UNITED STATES:

OUTLOOK: We continue to expect real GDP growth to be 1.5% (annual rate) in 4Q2010 and to remain below trend in 2011, largely owing to the limited political appetite to extend fiscal stimulus. We forecast 2011 annual growth of just 1.8%. The labour market should remain correspondingly weak, with unemployment averaging 9.7% in 2H2010 and 10% in 2011.
KEY ISSUES: US data has been weaker in recent months, supporting our view of a growth slowdown, although better recent developments suggest recession risk is receding. The
FOMC announced further quantitative easing at its November meeting, indicating that it plans to purchase $600bn of longer-term Treasuries by the end of 2Q2011.

JAPAN

OUTLOOK : We forecast real GDP growth of –2.2% (annual rate) in 4Q2010, resulting in 2010 annual growth of 3.5%. In 2011, we expect growth to slow to 0.9%. We believe that the strong growth rate of 3.9% in 3Q2010 was a temporary surge ahead of the expiry of government subsidies. We expect growth to remain sluggish owing both to a pullback in consumption as these subsidies end, and to a slowdown in export growth stemming from weakness in the US and the recent appreciation of the Yen.
KEY ISSUES: The BoJ recently announced a comprehensive easing program, but the measures are modest in size and we believe the caveats attached mean that the effect will be muted. We believe further easing will depend on conditions in both the real economy and markets. The government is also currently considering a supplementary budget of roughly ¥5trillion, which presents upside risks to our 2011 forecasts and may necessitate further forecast revisions once details are made available.

EUROPE

OUTLOOK : We continue to expect real GDP growth in 2010 to be 1.8%yoy in the EU-27 and 1.7%yoy in the Euro-zone. In 2011, we forecast growth of 2.2%yoy for the EU-27 and 1.9%yoy for the Euro-zone. Our forecast remains above consensus, but we see several downside risks, including further escalation of problems in the Eurozone periphery, different FX developments from what we currently forecast, a greater than expected drag on growth from fiscal consolidation, and a dampening effect on growth from financial sector deleveraging.
KEY ISSUES: Recent European data indicate an easing in momentum, confirming the start of the phase of moderation and stabilisation that we expect in 2H2010. Although peripheral spreads have widened recently and there has been increased speculation about an Ireland ‘rescue package’, we do not expect these to have major Euroland-wide spillover effects. Moreover, we do not expect sluggish growth in the US to have a significant growth impact in Europe, unless dramatic Euro appreciation results.

NON-JAPAN ASIA

OUTLOOK: For Asia ex Japan we forecast growth of 9.0% in 2010 and 8.4% in 2011. We expect growth in most countries across the region to slow in 2011, but remain at solid levels. In China, we forecast real GDP growth of 10.1% for 2010 and a return to its trend level of 10.0% for 2011.
KEY ISSUES : In China, recent activity data has been strong, underpinned by inventory increases and ‘stealth’ policy easing from strong fiscal spending and credit growth. We believe that strong domestic demand and rising food prices will continue to put upward pressure on inflation, which should push Chinese authorities to tighten policy in response.

LATIN AMERICA

OUTLOOK : Our LatAm growth forecast is 6.3% for 2010 and 4.6% in 2011. Our view is optimistic across most of the region thanks to accommodative monetary policy stances, strong domestic demand growth, strengthening labour markets and firming credit inflows.
KEY ISSUES: We expect real GDP growth in Brazil to accelerate to an above-trend rate of 7.5%yoy in 2010, due mainly to expansionary macro policies. The new administration of President-elect Dilma Rousseff will likely focus on boosting growth, with the government playing a more visible role.

CENTRAL & EASTERN EUROPE, MIDDLE EAST AND AFRICA

OUTLOOK: CEEMEA activity data has slowed in recent months on the back of slowing industrial momentum and increased uncertainty about global growth. Most of our forecasts remain above consensus. Economies with strong balance sheet structures and easy financial conditions, most notably Poland, Turkey and the Czech Republic, should outperform.
KEY ISSUES: We expect a moderation in growth across the region in 2H2010, driven mainly by weakening external demand. Despite this, monetary policy is currently still overly accommodative, and we expect central banks to withdraw excess stimulus gradually through a combination of rate hikes, currency appreciation and sterilised FX intervention.

CENTRAL BANK INTEREST RATE POLICIES

UNITED STATES: FOMC
CURRENT SITUATION: The Fed cut the funds rate to a range of 0%-0.25% on December 16, 2008.
NEXT MEETINGS: December 14 – January 26
EXPECTATION: We expect the Fed to keep the funds rate near 0% through the end of 2011.

JAPAN: BoJ Monetary Policy Board
CURRENT SITUATION: The BoJ cut the overnight call rate to a range of 0%-0.1% on October 5, 2010.
NEXT MEETINGS: December 21 – January 25
EXPECTATION: We expect the BoJ to keep the policy rate near 0% through the end of 2011.

EUROLAND: ECB Governing Council
CURRENT SITUATION: The ECB cut rates by 25bp to 1.0% on May 7, 2009.
NEXT MEETINGS: November 18 – December 2
EXPECTATION: We expect the ECB to keep the policy rate on hold until 3Q2011.

UK: BoE Monetary Policy Committee
CURRENT SITUATION: The BoE cut rates by 50bp to 0.5% on March 5, 2009.
NEXT MEETINGS: December 9 – January 13
EXPECTATION: We expect the BoE to keep the policy rate on hold until a 50bp hike in 4Q2011.

Source: ETFWorld – Goldman Sachs Global Economics, Commodities and Strategy Research

Normal 0 14 MicrosoftInternetExplorer4